venturi123 / DRLinFluids

An open-source Python platform of coupling deep reinforcement learning and OpenFOAM
Other
135 stars 28 forks source link

"Question consultation" or "inquiry." #19

Closed qzlkentomomota closed 4 months ago

qzlkentomomota commented 5 months ago

Hello, Blogger. Your work is amazing and I deeply admire what you have accomplished. I am a beginner and I would like to know why the reference area (Aref) in your square for the force coefficients in incompressible flow is set to 2.308194622e-3. Shouldn't it be equal to lref? Moreover, your reference length should theoretically be 0.05 if the Reynolds number is 100, right? Your explanation is very important to me, and I would greatly appreciate it. Thank you very much.

venturi123 commented 5 months ago

No, the value lRef=0.0508 is derived from many typical papers in simulations and experiments. You can refer to this paper and find a more similar setup in its references.

Regarding your second question, when handling a two-dimensional flow and calculating the force coefficients, it considers the height direction when integrating the force on our cylinder, if I remember correctly, even though the N-S equation does not account for it. If you check the mesh, you will find that multiplying the mesh height and reference length obtains Aref.

Moreover, the Reynolds number is related to inflow velocity, reference length, and kinematic viscosity. A common practice is to adjust the reference length and inflow velocity to appropriate values and keep them constant while changing the viscosity to adjust the Reynolds number to a predetermined value. Please refer to this paper for specific methods.

venturi123 commented 4 months ago

It seems there are no more questions. I'll close the issue now.